Filters remove water from slurry to produce a damp cake: concentrates for shipping, tailings for dry stacking, or leach residues for washing. Most filters draw water through a cloth under vacuum or force it through under pressure, leaving the solids on the cloth as a cake.
The filtration rate depends mainly on the cake. Fine, clay-rich solids form a dense cake of low permeability, whereas coarse, clean solids form an open cake that drains readily. A leaf test measures this behaviour on a small area of filter cloth, so that the filter can be sized before it is purchased.
Leaf tests are the standard bench test for vacuum filters, and the same principles underlie pressure filter tests. This article covers the procedure, the rate and moisture calculations, the filtration equation that describes cake resistance, and scale-up to a plant filter.
Leaf test procedure
- Prepare the slurry at the plant filter feed density, with the flocculant or filter aid that the plant will use.
- Set the vacuum and cycle times to match the filter type. For a disc or drum filter, the form time is the time the leaf would spend submerged and the dry time is the time out of the slurry before discharge.
- Form the cake by submerging the leaf (cloth side down for bottom-feed tests) under vacuum for the form time, with the slurry agitated, and record filtrate volume against time.
- Dry the cake by lifting the leaf and maintaining vacuum for the dry time, drawing air through the cake.
- Remove the cake, measure its thickness, weigh it wet, dry it and weigh it again, and check filtrate clarity.
- Repeat the test at several form and dry times to establish the relationship between rate and moisture.
Filtration rate and filter area
where
- dry cake mass from one leaf test (kg)
- test leaf area (m²)
- full cycle time: form + dry + discharge (h)
- dry solids to filter (kg/h)
Worked example: a 0.1 m² leaf with a form time of 1.0 min, a dry time of 2.0 min and 0.5 min allowed for discharge, giving a 3.5 min cycle, produces a dry cake of 1.20 kg. The rate is 1.20 ÷ (0.1 × 3.5 ÷ 60) = 206 kg of dry solids per m² per hour. A plant filtering 2,400 t/d (100,000 kg/h) of dry solids therefore requires 100,000 ÷ 206 = 486 m², or 608 m² of filter area with a scale-up allowance of 1.25 chosen for this example.
Cake moisture = (wet mass − dry mass) ÷ wet mass. A wet cake of 1.364 kg with a dry mass of 1.20 kg has 12% moisture. Longer dry times reduce moisture but lengthen the cycle and therefore reduce the rate, and the test series defines this trade-off.
The filtration equation
For a cake-forming filter at constant pressure, the time to collect a given volume of filtrate follows the classical filtration equation:
where
- ,
- time from the start of filtration (s) and filtrate volume collected (m³)
- filtrate viscosity (Pa·s; about 0.001 for water at 20 °C)
- specific cake resistance (m/kg)
- mass of dry cake formed per volume of filtrate (kg/m³)
- ,
- filter area (m²) and pressure difference across cake and cloth (Pa)
- resistance of the filter cloth (1/m)
A plot of t/V against V gives a straight line. The slope gives the specific cake resistance α, and the intercept gives the cloth resistance.
| Time t (s) | 10 | 20 | 30 | 45 | 60 |
|---|---|---|---|---|---|
| Filtrate V (L) | 0.415 | 0.592 | 0.729 | 0.896 | 1.037 |
| t/V (s/L) | 24.1 | 33.8 | 41.2 | 50.2 | 57.9 |
The line has a slope of 54.2 s/L², which is 5.42 × 10⁷ s/m⁶. With μ = 0.001 Pa·s, A = 0.1 m², ΔP = 60 kPa and c = 1,158 kg of dry cake per m³ of filtrate (a 50% solids slurry giving a 12% moisture cake):
α = slope × 2A²ΔP ÷ (μc) = 5.42 × 10⁷ × 2 × 0.01 × 60,000 ÷ (0.001 × 1,158) = 5.6 × 10¹⁰ m/kg.
Specific cake resistance is a property of the solids and their preparation. It allows flocculants, grinds and ore types to be compared, and the effect of pressure and cycle time on rate to be predicted without testing every combination. Many cakes are compressible, so α itself rises with pressure; tests at two or more pressures quantify the effect.
Scale-up
Leaf tests usually give higher rates than the plant achieves, because of cloth blinding, uneven cake formation and imperfect discharge. Designers therefore apply a scale-up allowance based on experience with the filter type.
The tests should cover the full range of feed, because fines content and clay have a large effect on rate. Cloth selection should also be checked in the tests, since the cloth must retain fines without blinding. Pressure filters and belt filters have their own test procedures, which suppliers carry out as part of equipment selection.
Glossary
- Leaf test
- A bench test using a small filter leaf to copy one cycle of a vacuum filter.
- Filter cake
- The layer of solids left on the cloth.
- Filtrate
- The liquid that passes through the cake and cloth.
- Form time
- The part of the cycle when the cake is built up under vacuum in the slurry.
- Dry time
- The part of the cycle when air is drawn through the cake to remove moisture.
- Specific cake resistance, α
- How strongly a cake resists flow per kilogram of cake per square metre (m/kg).
- Cake moisture
- Water as a percentage of the wet cake mass.
- Blinding
- Fines lodging in the filter cloth and reducing its permeability.
Frequently asked questions
How is filtration rate calculated from a leaf test?
Divide the dry cake mass by the leaf area and by the full cycle time (form, dry and discharge), giving kilograms of dry solids per square metre per hour.
How is filter area calculated?
Divide the dry solids rate by the filtration rate from the leaf test, then apply a scale-up allowance for the filter type.
What is specific cake resistance?
A measure of the resistance of the cake to liquid flow, obtained from the slope of a plot of t/V against V. Higher values indicate slower filtration.
Why does the plant filter achieve less than the leaf test predicted?
Cloth blinding, uneven cake, poor discharge, and changes in feed fines and flocculation all reduce plant rates, which is the reason for applying scale-up allowances.
Discuss your plant with a metallurgist
Tiegel is a Perth metallurgy company. We help with testwork, plant optimisation, metal accounting and investigations, and we build the Tiegel Suite software.
Talk to a metallurgistReferences and further reading
- Wakeman, R.J. and Tarleton, E.S. (2005). Solid/Liquid Separation: Principles of Industrial Filtration. Elsevier.
- Wills, B.A. and Finch, J.A. (2016). Wills' Mineral Processing Technology, 8th edition. Butterworth-Heinemann. Chapter 15, Dewatering.
Worked examples and charts marked illustrative use invented figures, not data from any operation.